August
15,
2003
1
RISK
SCREEN
ON
THE
USE
OF
SUBSTITUTES
FOR
OZONE­
DEPLETING
SUBSTANCES
REFRIGERATION
AND
AIR
CONDITIONING
End­
Use
Replacement:
HCFC­
123
in
Autocascade
Refrigeration
Systems
and
CFC­
113
and
CFC­
11
in
Industrial
Process
Refrigeration
and
Heat
Transfer
Systems
Refrigerant:
HFE­
7000
This
risk
screen
does
not
contain
Clean
Air
Act
(
CAA)
Confidential
Business
Information
(
CBI)
and,
therefore,
may
be
disclosed
to
the
public.

1.
INTRODUCTION
Stratospheric
ozone­
depleting
substances
(
ODS)
are
being
phased
out
of
production
in
response
to
a
series
of
diplomatic
and
legislative
efforts
that
have
taken
place
over
the
past
few
years,
including
the
Montreal
Protocol
and
the
Clean
Air
Act
Amendments
of
1990
(
CAAA).
The
U.
S.
Environmental
Protection
Agency
(
EPA),
as
authorized
by
Section
612
of
the
CAAA,
has
developed
a
program
to
evaluate
the
human
health
and
environmental
risks
posed
by
alternatives
to
ODS.
The
main
purpose
of
EPA's
program,
called
the
Significant
New
Alternatives
Policy
(
SNAP)
program,
is
to
identify
acceptable
and
unacceptable
substitutes
for
ODS
in
specific
end
uses.

EPA's
decision
on
the
acceptability
of
a
substitute
is
based
largely
on
the
findings
of
a
screening
assessment
of
potential
human
health
and
environmental
risks
posed
by
the
substitute
in
specific
applications.
EPA
has
already
screened
a
large
number
of
substitutes
in
many
end­
uses
within
all
of
the
major
ODS­
using
sectors,
including
refrigeration
and
air
conditioning;
solvent
cleaning;
foam
blowing;
aerosols;
fire
extinguishing;
adhesives,
coatings,
and
inks;
and
sterilization.
The
results
of
these
risk
screens
are
presented
in
a
series
of
background
documents
that
are
available
in
this
docket.

The
purpose
of
this
report
is
to
supplement
EPA's
background
document
on
the
refrigeration
and
air
conditioning
sector
1
(
hereinafter
referred
to
as
the
Background
Document)
by
adding
to
the
list
of
potential
substitutes
for
end
uses
of
HCFC­
123,
CFC­
113,
and
CFC­
11
in
the
refrigeration
and
air
conditioning
sector.
The
proposed
end­
use
applications
considered
in
this
analysis
include
industrial
process
refrigeration,
heat
transfer,
and
autocascade
refrigeration
systems.
The
specific
proposed
CFC/
HCFC
substitute
examined
in
this
report
is
shown
in
Table
1.
Many
of
the
methodologies
used
to
conduct
this
risk
screen
are
identical
to
those
described
in
the
Background
Document.
The
reader
is
referred
to
the
Background
Document
for
a
complete
discussion
of
these
methods
and
their
limitations.
In
addition,
some
parts
of
this
analysis
are
based
on
several
methodology
documents
that
are
included
as
attachments
to
this
report.

Section
2
of
this
report
summarizes
the
results
of
the
risk
screen
for
the
proposed
substitute
listed
in
Table
1.
The
remainder
of
the
report
is
organized
similarly
to
the
Background
Document:
Section
3
presents
the
toxicity
values
used
for
the
risk
screen,
Section
4
presents
the
results
of
the
atmospheric
assessment,
Sections
5
and
6
discuss
occupational
and
general
population
exposures,
and
Section
7
discusses
potential
increases
in
atmospheric
releases
of
volatile
organic
compounds
(
VOCs).

1
U.
S.
EPA.
1994.
SNAP
Technical
Background
Document:
Risk
Screen
on
the
Use
of
Substitutes
for
Class
I
Ozone­
depleting
Substances:
Refrigeration
and
Air
Conditioning.
Stratospheric
Protection
Division.
March,
1994.
August
15,
2003
2
TABLE
1.
COMPOSITION
OF
HFE­
7000
Trade
Name
Constituent
Chemical
Formula
CAS
No.
Percent
by
Weight
HFE­
7000
(
Hydrofluoroether,
HFE­
301,
L­
13791,
[
F­
12025],
T­
6903,
or
methylperfluoropropylether)
1,1,1,2,2,3,3­
heptafluoro­
3­
methoxy­
propane
or
1,1,1,2,2,3,3­
heptafluoropropyl
methyl
ether
C4H3F7O
375­
03­
1
>
991
1Note
that
less
than
1
percent
of
HFE­
7000
is
composed
of
impurities
that
are
reasonably
anticipated
to
be
present
when
manufactured
for
commercial
purposes.
These
impurities
are
not
thought
to
be
present
in
quantities
sufficient
to
pose
a
risk
to
humans
or
to
the
environment.

2.
SUMMARY
OF
RESULTS
HFE­
7000
is
recommended
for
SNAP
approval
for
the
proposed
end
uses.
No
significant
risks
to
workers
are
estimated
according
to
occupational
exposure
modeling,
and
general
population
exposure
to
the
substitute
is
expected
to
be
below
levels
of
concern
for
noncancer
risks.
Nevertheless,
adequate
ventilation
should
be
used
throughout
the
manufacturing
and
disposal
processes.
Additionally,
the
EPA's
risk
screen
indicates
that
the
use
of
the
proposed
substitute
and
its
constituents
will
be
less
harmful
to
the
atmosphere
than
the
continued
use
of
CFCs.
For
applications
of
this
and
all
other
refrigerants,
EPA
recommends
that
American
Society
of
Heating,
Refrigerating
and
Air­
Conditioning
Engineers
(
ASHRAE)
Standards
15
and
34
be
followed.

3.
TOXICITY
REFERENCE
VALUES
FOR
SUBSTITUTES
To
assess
potential
health
risks
from
exposure
to
this
substitute
for
ODS
in
the
refrigeration
and
air
conditioning
sector,
EPA
identified
the
relevant
toxicity
threshold
values,
including
available
acceptable
exposure
levels
(
AELs),
for
comparison
to
modeled
exposure
concentrations
for
different
scenarios.
For
the
occupational
exposure
analysis,
potential
risks
from
chronic
worker
exposure
were
evaluated
as
were
risks
to
the
general
population
from
exposure
to
ambient
air
releases.
The
toxicity
threshold
values
for
this
assessment
are
shown
in
Table
2.
EPA's
approach
for
identifying
or
developing
these
values
is
discussed
in
Chapter
3
of
the
Background
Document.

4.
ATMOSPHERIC
MODELING
This
section
presents
EPA's
assessment
of
the
potential
impact
of
each
substitute
on
ozone
depletion
and
global
warming.
EPA's
approach
for
the
atmospheric
assessment
is
discussed
in
detail
in
Chapter
4
of
the
Background
Document.
Table
3
presents
the
ozone­
depleting
potential
(
ODP),
global
warming
potential
(
GWP),
and
atmospheric
lifetime
(
ALT)
for
the
constituents
of
the
proposed
substitute.

The
results
of
the
atmospheric
assessment
indicate
that
HFE­
7000
has
an
ALT
of
4.9
years
and
an
ODP
of
0.
The
GWP
of
this
substitute
is
370.
The
substitute
is
substantially
less
harmful
to
the
ozone
layer
than
the
continued
use
of
CFCs.
August
15,
2003
3
TABLE
2.
TOXICITY
THRESHOLD
VALUES
Chemical
AEL
(
Long­
term
Exposure)
STEL
(
Short­
term
Exposure)
RfC
C4H3F7O
75
ppm
8
hr
TWAa
225
ppm
15
min
TWAb
2
ppmc
Note:
TWA
=
Time­
Weighted
Average;
RfC
=
Reference
Concentration.
a
See
Attachment
1,
ICF
(
2001a),
"
Recommendation
for
an
8­
hour
Acceptable
Exposure
Limit
for
methoxy
heptafluoropropane
(
T­
6903)"
b
Based
on
3
times
the
AEL.
c
See
Attachment
2,
ICF
(
2001b),
"
Recommendation
for
an
RfC
for
methoxy
heptafluoropropane
(
T­
6903)"

TABLE
3.
ODP,
GWP,
AND
ALT
FOR
REFRIGERATION
AND
AIR
CONDITIONING
SUBSTITUTES
SUBSTITUTE
ODP
100
Year
GWP
(
relative
to
CO2)
ALT
(
years)

HFE­
7000
0
370
4.9
Source:
3M
Company
(
2001a)

5.
OCCUPATIONAL
EXPOSURE
AND
RISK
SCREENING
ANALYSIS
This
section
presents
estimates
of
potential
occupational
exposures
to
the
proposed
substitute
during
manufacture,
maintenance,
and
disposal
of
refrigeration
equipment
in
the
refrigeration
and
air
conditioning
sector.
These
estimated
exposures
are
then
compared
to
occupational
exposure
limits
to
determine
whether
they
are
of
potential
concern.
The
methodology
and
results
are
summarized
briefly
in
Section
5.1.
Analysis
of
flammability
is
presented
in
Section
5.2.

5.1
Methodology
For
Industrial
Process
Refrigeration
and
Heat
Transfer
Systems
using
HFE­
7000,
occupational
exposures
are
not
likely
to
be
of
concern.
In
particular,
these
systems
would
not
be
likely
to
be
charged
in
a
typical
manufacturing
facility
and
certainly
would
not
pose
a
chronic
exposure
risk.
Large
systems
of
this
type
would
likely
be
designed
and
constructed
on­
site
on
an
as
needed
basis.
These
systems
would
then
be
charged
with
refrigerant
from
a
large
tank
truck
at
system
start­
up
and
would
only
need
to
be
serviced
once
or
twice
a
year.
Furthermore,
these
types
of
systems
would
be
initially
charged
or
maintained
by
experienced
personnel
using
proper
industrial
hygiene
techniques,
in
well­
ventilated
or
outdoor
locations.

For
the
other
proposed
end­
use,
Autocascade
Refrigeration
Systems,
although
manufacture
would
take
place
at
a
manufacturing
facility,
these
units
are
designed
and
built
to
meet
specific
customer
needs;
therefore,
the
units
would
not
be
produced
on
an
on­
going,
regular
schedule.
As
such,
occupational
exposure
is
also
not
a
concern
during
manufacture,
maintenance,
or
disposal
for
this
proposed
end­
use.
August
15,
2003
4
5.2
Flammability
Characteristics
It
is
important
to
consider
the
flammability
of
substances
when
investigating
their
acceptability
for
use
as
refrigerants
since
substitutes
that
are
flammable
could
pose
safety
concerns
to
workers.
However,
according
to
the
material
safety
data
sheet
(
MSDS)
provided
by
the
submitter,
HFE­
7000
is
nonflammable
and
thus
flammability
concerns
are
not
relevant
to
this
analysis.

6.
GENERAL
POPULATION
EXPOSURE
AND
RISK
SCREENING
ANALYSIS
This
section
screens
potential
risks
to
the
general
population
from
exposure
to
ambient
air
releases
of
the
substitute
examined
for
the
proposed
end­
uses.
For
two
of
the
end
uses
examined
in
this
risk
screen,
Industrial
Process
Refrigeration
and
Heat
Transfer
Systems
using
HFE­
7000,
the
installation
of
the
system
would
take
place
on­
site
and
only
once.
As
such,
HFE­
7000
does
not
pose
a
chronic
exposure
risk
to
the
general
population
in
these
proposed
end­
uses.
For
the
other
proposed
end­
use,
Autocascade
Refrigeration
Systems,
although
manufacture
would
take
place
at
a
manufacturing
facility,
these
units
are
highly
specialized
and
would
be
designed
and
built
to
meet
specific
customer
needs
on
an
as
needed
basis
(
i.
e.,
they
would
not
be
produced
on
an
on­
going,
regular
schedule)
and
also
would
not
pose
a
chronic
exposure
risk
to
the
general
population.
However,
if
in
the
future,
other
end­
uses
are
proposed
in
which
the
general
population
would
be
routinely
exposed
to
HFE­
7000
(
i.
e.,
exposures
resulting
from
releases
from
a
typical
manufacturing
facility),
exposures
should
be
compared
to
the
reference
concentration
(
RfC)
developed
for
HFE­
7000
(
see
Attachment
3).

7.
VOLATILE
ORGANIC
COMPOUND
ANALYSIS
With
respect
to
HFE­
7000,
the
analysis
presented
in
the
Background
Document
has
shown
that
potential
emissions
of
VOCs
from
all
substitutes
for
all
end
uses
in
the
refrigeration
and
air
conditioning
sector
are
likely
to
be
insignificant
relative
to
VOCs
from
all
other
sources
(
i.
e.,
both
anthropogenic
and
biogenic).

8.
REFERENCES
EPA.
1994.
"
SNAP
Technical
Background
Document:
Risk
Screen
on
the
Use
of
Substitutes
for
Class
I
Ozonedepleting
Substances:
Refrigeration
and
Air
Conditioning."
Stratospheric
Protection
Division.
March,
1994.

ICF.
1999.
"
Methodology
for
Determining
Zone
Estimates."
ICF
Incorporated.
April,
1999.

ICF.
2001a.
"
Recommendation
for
an
8­
hour
Acceptable
Exposure
Limit
for
methoxy
heptafluoropropane
(
T­
6903)."
Prepared
by
H.
Clewell
and
G.
Lawrence
under
EPA
Contract
No.
68­
D­
00­
266,
WA
0­
04,
Task
2.

ICF.
2001b.
"
Recommendation
for
an
RfC
for
methoxy
heptafluoropropane
(
T­
6903)
by
ICF
Consulting."
Prepared
by
H.
Clewell
and
G.
Lawrence
under
EPA
Contract
No.
68­
D­
00­
266,
WA
0­
04,
Task
2.

3M
Company
2001a.
Significant
New
Alternatives
Policy
(
SNAP)
Program
HFE­
7000
Submission
to
the
U.
S.
Environmental
Protection
Agency,
February
2001.

3M
Company
2001b.
Phone
conversation
with
Kurt
Werner
from
3M
Company,
ICF
Consulting,
and
EPA.
April
12.
August
15,
2003
5
ATTACHMENT
1.

ICF.
2001a.
"
Recommendation
for
an
8­
hour
Acceptable
Exposure
Limit
for
methoxy
heptafluoropropane
(
T­
6903)"
Prepared
by
ICF
under
EPA
Contract
No.
68­
D­
00­
266,
WA
2­
04,
Task
2.
Revised
March
20,
2003.

The
purpose
of
this
report
is
to
recommend
an
Acceptable
Exposure
Limit
(
AEL)
for
occupational
exposure
to
methoxy
heptafluoropropane
(
T­
6903).
The
toxicity
study
reviewed
for
this
chemical
was
a
30­
day
inhalation
toxicity
study
in
rats
(
Arts
et
al.
1999).

Recommended
AEL:
75
ppm
Basis:
Endpoint:
Decreased
body
weight
Study:
A
sub­
acute
(
30
day)
inhalation
toxicity
study,
including
a
recovery
study,
with
T­
6903
in
rats
Protocol:
Nose­
only
inhalation
exposure
for
6
hours/
day
5
days/
week
for
30
days
(
5/
sex/
group)
with
a
14­
day
recovery
period
in
satellite
groups
(
control
and
high­
concentration
only)

Doses:
0
(
air
control),
1,000,
10,000
or
30,000
ppm
NOAEL:
1,000
ppm
LOAEL:
10,000
ppm
HEC:
750
ppm
Uncertainty/
Modifying
Factors:

3
­
animal
to
human
extrapolation
3
­
use
of
a
less
than
90­
day
study
Total
=
10
Justification:

In
a
30­
day
nose­
only
inhalation
study,
Crl:(
WI)
WU
Wistar­
derived
rats
(
5/
sex/
group)
were
exposed
to
0
(
air
control),
1,000,
10,000
or
30,000
ppm
T­
6903
6
hours/
day,
5
days/
week
for
30
consecutive
days
(
Arts
et
al.
1999).
Additional
groups
of
5/
sex
in
the
control
and
high­
concentration
groups
were
allowed
a
14­
day
recovery
period.
The
endpoints
evaluated
were
clinical
signs
of
toxicity,
body
weights,
food
consumption
and
food
conversion
efficiency,
and
hematological,
clinical
chemistry
and
urinalysis
parameters.
At
necropsy,
the
major
organs
were
collected
and
weighed
and
preserved
for
microscopic
examination.
In
addition,
in
order
to
evaluate
potential
peroxisome
proliferation,
liver
samples
were
collected
and
acyl­
CoA
oxidase
and
lauric
acid
hydroxylase
activity
and
protein
content
were
measured.
August
15,
2003
6
The
concentration­
related
statistically­
significant
changes
reported
in
the
study
are
summarized
as
follows.
Decreases
in
mean
body
weights
in
the
mid­
and
high­
concentration
males
and
high­
concentration
females
that
were
accompanied
by
slight
(
not
statistically
significant)
decreases
in
food
consumption
and
food
conversion
efficiency
were
reported.
Mean
body
weights
remained
significantly
decreased
in
the
high­
concentration
males
and
females
at
7
days
after
termination
of
exposure,
but
were
comparable
to
the
controls
14
days
after
treatment.
Partial
thromboplastin
times
were
significantly
decreased
in
the
mid­
and
high­
concentration
females
at
the
end
of
treatment.
At
the
end
of
the
treatment
period,
statistically­
significant
increases
in
ALP
(
mid­
and
high­
concentration
males),
albumin/
globulin
ratio
(
all
treated
males
and
mid­
and
high­
concentration
females),
albumin
(
mid­
and
highconcentration
females),
triglycerides
(
mid­
and
high­
concentration
females)
and
phospholipids
(
all
treated
females)
were
reported.
Cholesterol
levels
were
significantly
decreased
in
all
treated
males.
However,
at
the
end
of
the
twoweek
recovery
period,
the
only
statistically­
significant
clinical
chemistry
change
was
an
increase
in
ALP
in
the
highconcentration
males.

Statistically­
significant
changes
in
urinalyses
parameters
at
the
end
of
treatment
were
increased
urine
volume
(
high­
concentration
males),
decreased
urinary
creatinine
(
high­
concentration
males
and
females),
and
in
all
treated
males
and
females,
increased
urinary
fluoride,
increased
urinary
fluoride
excretion/
16
hours
and
increased
urinary
fluoride
excretion/
mole
creatinine
was
reported.
At
the
end
of
the
recovery
period,
the
urinary
fluoride
excretion
parameters
in
the
high­
concentration
males
and
females
remained
significantly
increased,
when
compared
with
the
controls.
Lauric
acid
hydroxylase
activity
(
all
treated
males
and
high­
concentration
females)
and
acyl­
CoA
activity
(
all
treated
males
and
mid­
and
high­
concentration
females)
were
significantly
increased
at
the
end
of
treatment,
but
not
after
the
recovery
period.
At
necropsy,
significant
increases
in
absolute
and
relative
liver
weights
(
all
treated
males
and
high­
concentration
females)
were
reported.
Relative
liver
weights
were
also
significantly
increased
at
the
end
of
the
recovery
period.
A
significant
increase
in
relative
testes
weight
was
reported
in
the
highconcentration
males
at
the
end
of
treatment.
The
only
microscopic
lesion
with
a
statistically­
significantly
increased
incidence
was
hepatocellular
hypertrophy
(
all
treated
males
and
low­
and
mid­
concentration
females)
reported
at
the
end
of
treatment,
but
not
after
the
recovery
period.

The
results
of
Arts
et
al.
(
1999)
indicated
that
exposures
to
T­
6903
were
associated
with
decreased
body
weights,
alterations
in
lipid
metabolism,
increased
liver
and
testes
weights,
increases
in
the
incidence
of
hepatocellular
hypertrophy,
increased
lauric
acid
hydroxylase
activity
and
acyl­
CoA
activity
and
elevated
urine
fluoride
levels.
Increases
in
lauric
acid
hydroxylase
activity
and
acyl­
CoA
activity
are
considered
to
be
a
biomarker
for
peroxisome
proliferation,
which
in
rats
results
in
liver
hypertrophy,
increases
in
liver
weights
and
alterations
in
lipid
metabolism
(
Bentley
et
al.
1993).
However,
the
relevance
of
peroxisome
proliferation
and
increased
palmitoyl
CoA
activity
in
the
rat
is
questionable
with
regard
to
human
health.
Because
rodents
have
an
approximately
10­
fold
greater
concentration
of
PPAR

,
the
receptor
that
mediates
peroxisome
proliferation,
than
do
humans
(
Palmer
et
al.
1998),
rats
are
highly
sensitive
to
chemicals
that
activate
PPAR

.
In
contrast,
humans
have
been
reported
to
be
nonresponsive
to
chemicals
that
stimulate
PPAR

and
peroxisome
proliferation,
and
are
therefore
considered
to
be
less
sensitive
to
chemicals
at
levels
that
may
cause
effects
in
rats.
Therefore,
the
changes
in
liver
weights,
microscopic
hepatic
changes
and
the
increases
in
lauric
acid
hydroxylase
activity
and
acyl­
CoA
activity
reported
by
Arts
(
1999)
were
not
considered
to
be
relevant
to
human
health.

Guinea
pigs,
which
are
less
responsive
to
peroxisome
proliferators,
when
compared
with
rats
(
Lake
et
al.
2000),
exposed
to
10,000
ppm
T­
6903
6
hours/
day
for
10
days,
mean
body
weights
and
body
weight
gain
over
the
exposure
period
were
virtually
the
same
(
within
1­
2%)
in
treated
groups
and
in
the
controls
(
Lieder
2000).
This
suggests
that
the
decreased
body
weights
observed
in
the
mid­
and
high­
concentration
males
in
the
Arts
(
1999)
study
were
also
secondary
to
peroxisome
proliferation
effects.
Nevertheless,
this
can
not
be
stated
with
certainty
given
the
observed
decreases
in
food
consumption
in
this
case.
Therefore,
the
body
weight
changes
observed
in
male
rats
in
the
Arts
(
1999)
study
were
considered
the
critical
effect
and
the
NOAEL
for
this
study
was
1,000
ppm.

Dosimetric
Adjustments:

The
human
equivalent
concentration
(
HEC)
was
calculated
using
the
default
value
of
1
for
the
ratio
of
the
rat
and
human
blood:
air
partition
coefficient.
The
exposure
duration
of
6
hours/
day
was
adjusted
for
the
normal
8­
hour
occupational
exposure.
Because
the
exposure
of
interest
(
occupational)
is
5
days/
week,
no
additional
adjustments
were
made
for
days/
week.
The
resulting
HEC
is
750
ppm.
August
15,
2003
7
Uncertainty
Factors:

Because
the
RfC
dosimetry
guidelines
(
EPA
1994)
were
used
to
derive
the
HEC,
the
default
uncertainty
factor
for
differences
in
pharmacokinetics
was
not
applied.
Although
the
relevance
of
the
decreased
body
weights
reported
in
rats
in
the
Arts
(
1999)
study
is
questionable
with
regard
to
human
health,
it
is
possible
that
this
effect
was
not
part
of
the
peroxisome
proliferation
response.
Further,
data
from
a
90­
day
study
would
likely
allow
for
a
better
characterization
and
assessment
of
the
relevance
of
this
response.
Therefore,
an
uncertainty
factor
of
3
was
applied
for
potential
differences
in
pharmacodynamics
and
another
factor
of
3
was
applied
for
the
lack
of
a
subchronic
toxicity
study.
A
total
uncertainty
factor
of
10
results.
Application
of
the
total
uncertainty
factor
of
10
to
the
HEC
results
in
a
recommended
AEL
of
75
ppm.

Additional
Comments:

While
not
necessarily
an
indication
of
toxicity,
the
increase
in
urinary
fluoride
associated
with
exposures
to
T­
6903
at
all
levels
of
treatment
and
the
continued
elevation
of
urine
fluoride
levels
after
the
14­
day
recovery
period
is
a
cause
for
concern
regarding
human
occupational
exposure.
Specifically,
increases
in
fluoride
levels,
whether
associated
with
exposure
to
fluoride
or
from
metabolism
of
fluorinated
materials
such
as
T6903,
can
lead
to
a
condition
known
as
skeletal
fluorosis.
Due
to
differences
in
pharmacokinetics,
the
production
of
fluoride
by
metabolism
would
be
expected
to
decrease
relative
to
body
weight
in
the
human
as
compared
to
the
rat;
however,
the
clearance
of
fluoride
into
the
urine
would
decrease
similarly.
Therefore,
steady­
state
urinary
fluoride
levels
associated
with
inhalation
of
T­
6903
would
normally
be
similar
in
humans
and
rats.

The
concentration
of
urinary
fluoride
resulting
from
daily
exposure
to
1,000
ppm
T­
6903,
the
lowest
concentration
tested,
for
6
hours
was
24
and
14
mg/
L
in
male
and
female
rats,
respectively,
which
is
approximately
a
factor
of
3
greater
than
the
range
of
the
BEIs
for
urinary
fluoride
in
workers
recommended
by
the
ACGIH:
3
mg/
L
(
or
3
mg/
g
creatinine)
for
urine
collected
prior
to
the
workshift,
and
10
mg/
L
(
or
10
mg/
g
creatinine)
for
urine
collected
at
the
end
of
the
shift.
Using
an
uncertainty
factor
of
3
would
thus
result
in
an
exposure
that
would
likely
result
in
urine
fluoride
levels
within
the
range
of
BEIs.
However,
in
the
rat,
the
induction
of
hepatic
microsomal
enzymes
is
frequently
observed
with
peroxisome
proliferators.
Enzyme
induction
would
result
in
increased
metabolism
of
the
parent
compound
and
subsequent
release
of
fluoride
into
the
blood
in
the
rat.
Because
humans
are
less
responsive,
if
responsive
at
all,
to
the
effects
of
peroxisome
proliferators,
it
is
likely
that
urine
fluoride
levels
in
workers
exposed
to
comparable
levels
of
T­
6903
would
actually
be
less
than
in
the
rat,
i.
e.,
below
the
ACGIH
recommendations.
This
is
supported
by
data
collected
from
guinea
pigs
exposed
to
a
much
higher
concentration
(
10,000
ppm)
of
T­
6903
(
Lieder
2000).
After
a
single
6­
hour
exposure,
mean
urine
fluoride
levels
were
approximately
4.5­
fold
higher
than
in
the
unexposed
controls.
Even
after
10
exposures
(
6
hours/
day)
to
10,000
ppm
T­
6903,
urine
fluoride
levels
were
only
7­
fold
higher
than
the
controls.
This
suggests
that
steady­
state
had
been
reached
and
enzyme
induction
did
not
occur
in
guinea
pigs.
Therefore,
no
additional
uncertainty
factor
was
applied
for
fluoride
production.
However,
it
is
recommended
that
urine
fluoride
levels
be
monitored
in
exposed
workers
in
order
to
ensure
that
urine
fluoride
levels
remain
below
the
levels
recommended
by
the
ACGIH.
August
15,
2003
8
References
Arts
JH,
Kuper
CF,
and
Muijser
H.
(
1999).
A
sub­
acute
(
30
day)
inhalation
toxicity
study,
including
a
recovery
study,
with
T­
6903
in
rats.
TNO
Report
No.
V99.566.
TNO
Nutrition
and
Food
Research
Institute,
The
Netherlands.

Bentley
P,
Calder
I,
Elcombe
C,
et
al.
1993.
Hepatic
peroxisome
proliferation
in
rodents
and
its
significance
for
humans.
Fd.
Chem.
Toxic.
31:
(
11).
pp.
857­
907.

Lake
BG,
Rumsby
PC,
Price
RJ,
Cunningham
ME.
2000.
Species
differences
in
hepatic
peroxisome
proliferation,
cell
replication
and
transforming
growth
factor­
beta
1
gene
expression
in
the
rat,
Syrian
hamster
and
guinea
pig.
Mutat.
Res.
Mar
17;
448(
2):
213­
25.

Lieder
2000.
Unpublished
data.

Palmer,
CN,
Hsu
M­
H,
Griffin,
KJ,
Raucy
JL,
Johnson
EF.
1998.
Peroxisome
proliferator
activated
receptor
­

expression
in
human
liver.
Mol.
Pharmacol.
53:
14­
22.

EPA.
1994.
Methods
for
Derivation
of
Inhalation
Reference
Concentrations
and
Application
of
Inhalation
Dosimetry.
EPA/
600/
8­
90/
066F.
Office
of
Research
and
Development,
Washington,
DC.
October
1994.
August
15,
2003
9
ATTACHMENT
2.

ICF.
2001b.
"
Recommendation
for
an
RfC
for
methoxy
heptafluoropropane
(
T­
6903)
by
ICF
Consulting."
Prepared
by
ICF
under
EPA
Contract
No.
68­
D­
00­
266,
WA
2­
04,
Task
2.
Revised
March
20,
2003
The
purpose
of
this
report
is
to
recommend
a
reference
concentration
(
RfC)
for
occupational
exposure
to
methoxy
heptafluoropropane
(
T­
6903).
The
toxicity
study
reviewed
for
this
chemical
was
a
30­
day
inhalation
toxicity
study
in
rats
(
Arts
et
al.
1999).

Recommended
RfC:
2
ppm
Basis:
Endpoint:
Decreased
body
weight
Study:
A
sub­
acute
(
30
day)
inhalation
toxicity
study,
including
a
recovery
study,
with
T­
6903
in
rats
Protocol:
Nose­
only
inhalation
exposure
for
6
hours/
day
5
days/
week
for
30
days
(
5/
sex/
group)
with
a
14­
day
recovery
period
in
satellite
groups
(
control
and
high­
concentration
only)

Doses:
0
(
air
control),
1,000,
10,000
or
30,000
ppm
NOAEL:
1,000
ppm
LOAEL:
10,000
ppm
HEC:
179
ppm
Uncertainty/
Modifying
Factors:

3
­
animal
to
human
extrapolation
3
­
human
variability
10
­
study
duration
and
database
limitations
Total
=
100
Justification:

In
a
30­
day
nose­
only
inhalation
study,
Crl:(
WI)
WU
Wistar­
derived
rats
(
5/
sex/
group)
were
exposed
to
0
(
air
control),
1,000,
10,000
or
30,000
ppm
T­
6903
6
hours/
day,
5
days/
week
for
30
consecutive
days
(
Arts
et
al.
1999).
Additional
groups
of
5/
sex
in
the
control
and
high­
concentration
groups
were
allowed
a
14
day
recovery
period.
The
endpoints
evaluated
were
clinical
signs
of
toxicity,
body
weights,
food
consumption
and
food
conversion
efficiency,
and
hematological,
clinical
chemistry
and
urinalysis
parameters.
At
necropsy,
the
major
organs
were
collected
and
weighed
and
preserved
for
microscopic
examination.
In
addition,
in
order
to
evaluate
potential
peroxisome
proliferation,
liver
samples
were
collected
and
acyl­
CoA
oxidase
and
lauric
acid
hydroxylase
activity
and
protein
content
were
measured.

The
concentration­
related
statistically­
significant
changes
reported
in
the
study
are
summarized
as
follows.
Decreases
in
mean
body
weights
in
the
mid­
and
high­
concentration
males
and
high­
concentration
females
that
were
accompanied
by
slight
(
not
statistically
significant)
decreases
in
food
consumption
and
food
conversion
efficiency
were
reported.
Mean
body
weights
remained
significantly
decreased
in
the
high­
concentration
males
and
females
at
7
days
after
termination
of
exposure,
but
were
comparable
to
the
controls
14
after
treatment.
Partial
thromboplastin
times
were
significantly
decreased
in
the
mid­
and
high­
concentration
females
at
the
end
of
treatment.
At
the
end
of
the
treatment
period,
statistically­
significant
increases
in
ALP
(
mid­
and
high­
concentration
males),
albumin/
globulin
ratio
(
all
treated
males
and
(
mid­
and
high­
concentration
females),
albumin
(
mid­
and
high­
August
15,
2003
concentration
females),
triglycerides
(
mid­
and
high­
concentration
females)
and
phospholipids
(
all
treated
females)
were
reported.
Cholesterol
levels
were
significantly
decreased
in
all
treated
males.
However,
at
the
end
of
the
twoweek
recovery
period,
the
only
statistically­
significant
clinical
chemistry
change
was
an
increase
in
ALP
in
the
highconcentration
males.

Statistically
­
ignificant
changes
in
urinalyses
parameters
at
the
end
of
treatment
were
increased
urine
volume
(
high­
concentration
males),
decreased
urinary
creatinine
(
high­
concentration
males
and
females),
and
in
all
treated
males
and
females,
increased
urinary
fluoride,
increased
urinary
fluoride
excretion/
16
hours
and
increased
urinary
fluoride
excretion/
mole
creatinine
was
reported.
At
the
end
of
the
recovery
period,
the
urinary
fluoride
excretion
parameters
in
the
high­
concentration
males
and
females
remained
significantly
increased,
when
compared
with
the
controls.
Lauric
acid
hydroxylase
activity
(
all
treated
males
and
high­
concentration
females)
and
acyl­
CoA
activity
(
all
treated
males
and
mid­
and
high­
concentration
females)
were
significantly
increased
at
the
end
of
treatment,
but
not
after
the
recovery
period.
At
necropsy,
significant
increases
in
absolute
and
relative
liver
weights
(
all
treated
males
and
high­
concentration
females)
were
reported.
Relative
liver
weights
were
also
significantly
increased
at
the
end
of
the
recovery
period.
A
significant
increase
in
relative
testes
weight
was
reported
in
the
highconcentration
males
at
the
end
of
treatment.
The
only
microscopic
lesion
with
a
statistically­
significantly
increased
incidence
was
hepatocellular
hypertrophy
(
all
treated
males
and
low­
and
mid­
concentration
females)
reported
at
the
end
of
treatment,
but
not
after
the
recovery
period.

The
results
of
Arts
et
al.
(
1999)
indicated
that
exposures
to
T­
6903
were
associated
with
decreased
body
weights,
alterations
in
lipid
metabolism,
increased
liver
and
testes
weights,
increases
in
the
incidence
of
hepatocellular
hypertrophy,
increased
lauric
acid
hydroxylase
activity
and
acyl­
CoA
activity
and
elevated
urine
fluoride
levels.
Increases
in
lauric
acid
hydroxylase
activity
and
acyl­
CoA
activity
are
considered
to
be
a
biomarker
for
peroxisome
proliferation,
which
in
rats
results
in
liver
hypertrophy,
increases
in
liver
weights
and
alterations
in
lipid
metabolism
(
Bentley
et
al.
1993).
However,
the
relevance
of
peroxisome
proliferation
and
increased
palmitoyl
CoA
activity
in
the
rat
is
questionable
with
regard
to
human
health.
Because
rodents
have
an
approximately
10­
fold
greater
concentration
of
PPAR

,
the
receptor
that
mediates
peroxisome
proliferation,
than
do
humans
(
Palmer
et
al.
1998),
rats
are
highly
sensitive
to
chemicals
that
activate
PPAR

.
In
contrast,
humans
have
been
reported
to
be
nonresponsive
to
chemicals
that
stimulate
PPAR

and
peroxisome
proliferation,
and
are
therefore
considered
to
be
less
sensitive
to
chemicals
at
levels
that
may
cause
effects
in
rats.
Therefore,
the
changes
in
liver
weights,
microscopic
hepatic
changes
and
the
increases
in
lauric
acid
hydroxylase
activity
and
acyl­
CoA
activity
reported
by
Arts
(
1999)
were
not
considered
to
be
relevant
to
human
health.

Guinea
pigs,
which
are
less
responsive
to
peroxisome
proliferators,
when
compared
with
rats
(
Lake
et
al.
2000),
exposed
to
10,000
ppm
T­
6903
6
hours/
day
for
10
days,
mean
body
weights
and
body
weight
gain
over
the
exposure
period
were
virtually
the
same
(
within
1­
2%)
in
treated
groups
and
in
the
controls
(
Lieder
2000).
This
suggests
that
the
decreased
body
weights
observed
in
the
mid­
and
high­
concentration
males
in
the
Arts
(
1999)
study
were
also
secondary
to
peroxisome
proliferation
effects.
Nevertheless,
this
can
not
be
stated
with
certainty
given
the
observed
decreases
in
food
consumption
in
this
case.
Therefore,
the
body
weight
changes
observed
in
male
rats
in
the
Arts
(
1999)
were
considered
the
critical
effect
and
the
NOAEL
for
this
study
was
1,000
ppm.

Dosimetric
Adjustments:

The
human
equivalent
concentration
(
HEC)
was
calculated
using
the
default
value
of
1
for
the
ratio
of
the
rat
and
human
blood:
air
partition
coefficient.
The
exposure
duration
of
6
hours/
day
was
adjusted
for
continuous
(
24
hours/
day)
exposure.
Because
the
exposure
of
interest
(
occupational)
is
5
days/
week,
an
adjustment
was
made
for
days/
week.

HEC
=
NOAEL
(
1000
ppm)
×
6
hours/
24
hours
×
5
days/
7days
The
resulting
HEC
was
179
ppm.

Uncertainty
Factors:
September
10,
2003
11
Because
the
RfC
dosimetry
guidelines
(
EPA
1994)
were
used
to
derive
the
HEC,
the
default
uncertainty
factor
for
differences
in
pharmacokinetics
was
not
applied.
A
factor
of
3
was
applied
for
differences
in
pharmacodynamics.

Based
on
the
available
toxicity
information,
there
is
no
evidence
that
a
particular
group
of
individuals
would
be
more
sensitive
to
the
effects
of
HFE
7000.
The
toxicity
observed
in
the
rat
study
was
likely
related
to
peroxisome
proliferation.
Rats
are
sensitive
to
the
effects
of
peroxisome
proliferation.
Consequently,
an
uncertainty
factor
of
3,
rather
than
a
full
factor
of
10,
was
applied
for
human
variability.

Although
the
relevance
of
the
decreased
body
weights
reported
in
rats
in
the
Arts
(
1999)
study
is
questionable
with
regard
to
human
health,
it
is
possible
that
this
effect
was
not
part
of
the
peroxisome
proliferation
response.
Moreover,
a
long­
term
(
e.
g.,
subchronic
or
chronic)
toxicity
study
was
not
available
and
the
potential
effects
of
exposures
to
HFE
7000
of
longer
duration
could
not
be
determined.
Therefore,
an
uncertainty
factor
of
10
was
applied
for
study
duration
and
for
database
limitations.

A
total
uncertainty
factor
of
100
results.
Application
of
the
total
uncertainty
factor
of
100
to
the
HEC
results
in
an
RfC
of
2
ppm.

Additional
Comments:

While
not
necessarily
an
indication
of
toxicity,
the
increase
in
urinary
fluoride
associated
with
exposures
to
T­
6903
at
all
levels
of
treatment
and
the
continued
elevation
of
urine
fluoride
levels
after
the
14­
day
recovery
period
is
a
cause
for
concern
regarding
human
occupational
exposure.
Specifically,
increases
in
fluoride
levels,
whether
associated
with
exposure
to
fluoride
or
from
metabolism
of
fluorinated
materials
such
as
T6903,
can
lead
to
a
condition
known
as
skeletal
fluorosis.
Due
to
differences
in
pharmacokinetics,
the
production
of
fluoride
by
metabolism
would
be
expected
to
decrease
relative
to
body
weight
in
the
human
as
compared
to
the
rat;
however,
the
clearance
of
fluoride
into
the
urine
would
decrease
similarly.
Therefore,
steady­
state
urinary
fluoride
levels
associated
with
inhalation
of
T­
6903
would
normally
be
similar
in
humans
and
rats.

The
concentration
of
urinary
fluoride
resulting
from
daily
exposure
to
1,000
ppm
T­
6903,
the
lowest
concentration
tested,
for
6
hours
was
24
and
14
mg/
L
in
male
and
female
rats,
respectively,
which
is
approximately
a
factor
of
3
greater
than
the
range
of
the
BEIs
for
urinary
fluoride
in
workers
recommended
by
the
ACGIH:
3
mg/
L
(
or
3
mg/
g
creatinine)
for
urine
collected
prior
to
the
workshift,
and
10
mg/
L
(
or
10
mg/
g
creatinine)
for
urine
collected
at
the
end
of
the
shift.
Using
an
uncertainty
factor
of
3
would
thus
result
in
an
exposure
that
would
likely
result
in
urine
fluoride
levels
within
the
range
of
BEIs.
However,
in
the
rat,
the
induction
of
hepatic
microsomal
enzymes
is
frequently
observed
with
peroxisome
proliferators.
Enzyme
induction
would
result
in
increased
metabolism
of
the
parent
compound
and
subsequent
release
of
fluoride
into
the
blood
in
the
rat.
Because
humans
are
less
responsive,
if
responsive
at
all,
to
the
effects
of
peroxisome
proliferators,
it
is
likely
that
urine
fluoride
levels
in
workers
exposed
to
comparable
levels
of
T­
6903
would
actually
be
less
than
in
the
rat,
i.
e.,
below
the
ACGIH
recommendations.
This
is
supported
by
data
collected
from
guinea
pigs
exposed
to
a
much
higher
concentration
(
10,000
ppm)
of
T­
6903
(
Lieder
2000).
After
a
single
6
hour
exposure,
mean
urine
fluoride
levels
were
approximately
4.5­
fold
higher
than
in
the
unexposed
controls.
Even
after
10
exposures
(
6
hours/
day)
to
10,000
ppm
T­
6903,
urine
fluoride
levels
were
only
7­
fold
higher
than
the
controls.
This
suggests
that
steady­
state
had
been
reached
and
enzyme
induction
did
not
occur
in
guinea
pigs.
Therefore,
no
additional
uncertainty
factor
was
applied
for
fluoride
production.
However,
it
is
recommended
that
urine
fluoride
levels
be
monitored
in
exposed
workers
in
order
to
ensure
that
urine
fluoride
levels
remain
below
the
levels
recommended
by
the
ACGIH.
September
10,
2003
12
References
Arts
JH,
Kuper
CF,
and
Muijser
H.
(
1999).
A
sub­
acute
(
30
day)
inhalation
toxicity
study,
including
a
recovery
study,
with
T­
6903
in
rats.
TNO
Report
No.
V99.566.
TNO
Nutrition
and
Food
Research
Institute,
The
Netherlands.

Bentley
P,
Calder
I,
Elcombe
C,
et
al.
1993.
Hepatic
peroxisome
proliferation
in
rodents
and
its
significance
for
humans.
Fd.
Chem.
Toxic.
31:
(
11).
pp.
857­
907.

Lake
BG,
Rumsby
PC,
Price
RJ,
Cunningham
ME.
2000.
Species
differences
in
hepatic
peroxisome
proliferation,
cell
replication
and
transforming
growth
factor­
beta
1
gene
expression
in
the
rat,
Syrian
hamster
and
guinea
pig.
Mutat.
Res.
Mar
17;
448(
2):
213­
25.

Lieder
2000.
Unpublished
data.

Palmer,
CN,
Hsu
M­
H,
Griffin,
KJ,
Raucy
JL,
Johnson
EF.
1998.
Peroxisome
proliferator
activated
receptor
­

expression
in
human
liver.
Mol.
Pharmacol.
53:
14­
22.

EPA.
1994.
Methods
for
Derivation
of
Inhalation
Reference
Concentrations
and
Application
of
Inhalation
Dosimetry.
EPA/
600/
8­
90/
066F.
Office
of
Research
and
Development,
Washington,
DC.
October
1994.
September
10,
2003
13
ATTACHMENT
3.

ICF.
1999.
"
Methodology
for
Determining
Zone
Estimates."
ICF
Incorporated.
April,
1999.

This
section
presents
the
methodology
used
to
determine
zone
estimates
for
each
end­
use
application
and
scenario
(
i.
e.,
manufacturing
or
disposal).
The
results
of
the
analysis,
including
estimated
exposure
concentrations,
are
presented
in
EPA
(
1999).

The
zone
estimates
used
for
the
EPA
(
1999)
analysis
were
developed
based
on
estimates
presented
in
the
EPA
(
1994)
background
document
and
recent
investigations
of
facility
size.
ICF
obtained
estimates
of
the
square
footage
and
ceiling
height
of
typical
machine
rooms,
supermarkets
and
other
retail
food
facilities,
and
factories.
Table
1
presents
these
data.
These
estimates
were
then
used
to
determine
the
measurements
of
Zones
1,
2,
and
3
for
both
manufacture
and
disposal
operations
in
each
end­
use,
based
on
the
methodology
presented
in
EPA
(
1994).
Table
2
presents
these
zone
estimates.

The
default
zones
in
Table
2
were
taken
directly
from
EPA
(
1994).
For
the
non­
default
zones:
Zone
1
represents
a
space
around
an
emission
source
measuring
5'
x
5'
x
5'.
This
zone
is
larger
than
the
default
to
account
for
the
larger
equipment
sizes
in
the
chillers,
industrial
process,
retail
food,
and
cold
storage
end
uses.
Zone
2
represents
a
departure
from
the
default
to
account
for
the
estimated
ceiling
height
of
the
facility.
Zone
3
was
estimated
by
multiplying
the
square
footage
of
the
facility
by
the
ceiling
height.

Table
1.
Summary
of
Facility
Size
Estimates
Type
of
Facility
Square
Footage
End­
Use
Machine
Room
400
Chillers,
cold
storage,
and
industrial
process
Large
Retail
Food
Facility
5,000
Supermarket
back
room
Factory
10,000a
Default
factory
for
manufacture
of
all
equipment
a
Factories
can
be
much
significantly
larger
(
up
to
1,000,000
sq.
ft.).
This
represents
an
estimate
of
the
area
that
could
be
impacted
by
refrigerant
release.
August
15,
2003
16
Table
2.
Zone
Volume
Estimates
by
End­
Use
(
m3)

Manufacture
Disposal
End­
Use
Zone
1
Zone
2
Zone
3
Exposure
Location
Zone
1
Zone
2
Zone
3
Exposure
Location
Chillersa
4
82
5,667
Factory
4
61
170
On­
site,
machine
room
Industrial
Process
Refrigeration
4
82
5,667
Factory
4
61
170
On­
site,
machine
room
Cold
Storage
4
82
5,667
Factory
4
61
170
On­
site,
machine
room
Refrigerated
Transport
4
82
5,667
Factory
1
49
765
Default
Retail
Food
(
Supermarkets)
4
82
5,667
Factory
4
61
2,125
On­
site,
retail
food
Retail
Food
(
Stand
Alone
Cases)
4
82
5,667
Factory
1
49
765
Default
Motor
Vehicle
Air
Conditioning
4
82
5,667
Factory
1
49
765
Default
Household
Refrigeration
4
82
5,667
Factory
1
49
765
Default
Refrigerated
Appliancesb
4
82
5,667
Factory
1
49
765
Default
Residential
Unitary
AC
4
82
5,667
Factory
1
49
765
Default
Commercial
Unitary
AC
4
82
5,667
Factory
1
49
765
Default
Window
Unit
AC
4
82
5,667
Factory
1
49
765
Default
a
Includes
industrial
process
air
conditioning.

b
Includes
dehumidifiers,
ice
makers,
and
vending
machines.
